Temporal Difference Learning Methods for Control

By Rohit Pardasani

---

SARSA: Generalized Policy Iteration (GPI) with Temporal Difference (TD) ?

---

Q-Learning

---

Expected SARSA

---

How Expected SARSA does Off-Policy Learning

References

https://mitpress.mit.edu/books/reinforcement-learning
https://www.coursera.org/specializations/reinforcement-learning